PROJECT SUMMARY Nonalcoholic steatohepatitis (NASH) is an established precursor of end-stage liver disease and hepatocellular carcinoma. To date, liver biopsy remains the only test available to detect the two diagnostic features of NASH: hepatocyte ballooning and lobular inflammation, but it is limited by invasiveness and potential complications. The central hypothesis of this application is that magnetic resonance elastogram (MRE) can discriminate NASH from simple steatosis. The hypothesis is supported by strong preliminary data which show that with novel technological modifications, MRE can isolate imaging parameters that correlate with inflammation and ballooning. A single multiparametric test that combines inflammation and ballooning assessment by MRE, and steatosis measurement by MRI-proton density fat fraction has the potential to provide a comprehensive estimation of the 3 components of histologic NAFLD activity score (NAS) in one setting. The objective of this study is to determine and validate the diagnostic performance of this single multiparametric test (henceforth named Hepatogram) for the detection of NASH in human subjects. We will test the central hypothesis in 3 AIMs. In AIM 1 we will use a cohort of obese subjects scheduled for bariatric surgery to assess NASH histologically by liver biopsy obtained during bariatric surgery and noninvasively before surgery by Hepatogram. The imaging parameters of inflammation, ballooning and steatosis will be used to develop a statistical model to predict histologic NAS score. In AIM 2 we will validate this statistical model in an independent, non-bariatric NAFLD cohort, in whom NASH will be diagnosed by liver biopsy and assessed by Hepatogram. In AIM 3 we will examine the longitudinal diagnostic performance of Hepatogram in the detection of NASH regression after weight-loss interventions. We will reevaluate the bariatric cohort described in Aim 1, one year after bariatric surgery by assessing for NASH regression by Hepatogram and liver biopsy. The predictive performance of the statistical model developed in Aim 1 and validated in Aim 2 will be examined by comparing the predicted MRE-based change in NAS score to the change in histologic NAS. This proposal will apply technically innovative approaches to a large, rigorously characterized population with NAFLD to test the hypothesis that Hepatogram is an accurate imaging biomarker for NASH diagnosis, monitoring and response to interventions. Obviating the impractical need for liver biopsy to diagnose NASH before onset of fibrosis would positively impact the care of NAFLD patients. This work will be performed in an academically nurturing environment within Mayo Clinic. The candidate will be guided by a strong mentorship committee of experts in the areas that the candidate identified for necessary training. The candidate acquired a strong foundation in statistical skills in assessment of disease risk, prognostication and outcomes evaluation during her post- doctoral training. She now proposes to solidify her foundation and acquire a set of new and complementary skills necessary for an independent career in clinical research in nonalcoholic fatty liver disease.